Presentation on theme: "The Survey of Income and Program Participation (SIPP) * Using Topical Modules H. Luke Shaefer University of Michigan School of Social Work National Poverty."— Presentation transcript:
The Survey of Income and Program Participation (SIPP) * Using Topical Modules H. Luke Shaefer University of Michigan School of Social Work National Poverty Center This presentation is part of the NSF ‑ Census Research Network project of the Institute for Social Research at the University of Michigan. It is funded by National Science Foundation Grant No. SES 1131500.
Topical Modules: An Intro Most SIPP waves include extra topical modules with an additional bundle of questions Some of these topical modules are widely used: Adult Well-Being/Material Hardship measures are considered the best available by many Asset and liabilities data are considered very strong Some have hardly/never been used The reliability of these measures may or may not have been validated If you enter the wild west of an uncharted topical module (TM), look for ways to benchmark your estimates
Topical Modules: An Intro The schedule of these topical modules is available here: http://www.census.gov/programs-surveys/sipp/tech- documentation/topical-modules.html http://www.census.gov/programs-surveys/sipp/tech- documentation/topical-modules.html Topical modules come in a separate file, but take the same form as core wave files: person-month form Often TM measures are at the household level (such as assets/ material hardship), but the variables are duplicated in each household member’s record In some cases, the universe of the TM excludes some SIPP respondents You can merge topical module (TM) variables into your core files using the person identifier and wave TM observations generally attach to the 4 th reference month of the wave they were conducted in While they attach to this observation, TM questions vary considerably in terms of the reference period they cover
Merging Topical Modules T PersonWaveRef Month Wor k Insured Luke4100 4200 4310 4411 LeBron4111 4211 4311 4411 PersonWaveNet Worth Luke4A little LeBron4A lot -------------------------Core------------------------ ---- ---------Topical Module----------
Merging Topical Modules PersonWav e Ref Month WorkInsuredNet Worth Luke4100. 4200. 4310. 4411A little LeBron4111. 4211. 4311. 4411A lot PersonWaveNet Worth Luke4A little LeBron4A lot STATA SYNTAX (After generating a variable srefmon == 4 in the TM file. Also, make sure epppnum is in the right units!) Now load in your core data Merge 1:1 ssuid epppnum swave srefmon using sipp08t4.dta, keepusing(networth) Must create a variable for the reference month and set it equal to 4 gen srefmon = 4
Merging Topical Modules PersonWav e Ref Month WorkInsuredFood Secure Luke4100. 4200. 4310. 44111 LeBron4111. 4211. 4311. 44111 PersonWaveFood Secure Luke41 LeBron41 The SIPP’s food security questions have a four-month reference period, so they can be thought of as pertaining to the 4 months of the partner wave STATA SYNTAX Merge 1:1 ssuid epppnum swave srefmon using sipp08t4.dta, keepusing(foodsecure)
Food Security in the SIPP In the SIPP, a household is defined as being food insecure if they report at least two of the following, in reference to the previous 4 months (Nord, 2006). They are considered to have very low food security if they report at least 4. EAFLAST: The food the household bought didn’t last and they didn’t have money to get more (answers “often” or “sometimes”). EAFBALN: The household couldn’t afford to eat balanced meals (answers “often” or “sometimes”). EAFSKIP: The adults in the household ever cut the size of their meals or skipped meals because there wasn’t enough money for food (answer “yes”). EAFLESS: The adults in the household ever ate less than they felt they should because there wasn’t enough money to buy food (answer “yes”). EAFDAY: The adults in the household ever did not eat for a whole day because there wasn’t enough money for food (answer “yes”).
Shaefer & Gutierrez, 2013
Good Resource on Assets and Liabilities Data Czajka, J. L., Jacobson, J. E., & Cody, S. (2003). Survey estimates of wealth: A comparative analysis and review of the survey of income and program participation. Washington, DC: Mathematica Policy Research. Available at www.ssa.gov/policy/docs/contractreports/SurveyEstimatesWealth.pdf. SIPP has lower estimates of aggregate wealth and net worth. This appears to be related to: Underestimation of assets of the wealthy (as with income)—this accounts for 72% of the difference Assets not measured by the SIPP Other SIPP is MUCH better at estimating liabilities Measures of the value of family’s own home are very strong Good at measuring the value of cars
UNSECURED DEBT OF SINGLE WOMEN Shaefer, Song & William-Shanks, 2013
Based on Buchmueller, Orzel & Shore-Sheppard, 2014